Intelligent tutoring system

• US 6,807,535 B2
• Filed: 03/08/2001
• Issued: 10/19/2004
• Est. Priority Date: 03/08/2000
• Status: Expired due to Fees

• Current Assignee: LNK Corporation
• Original Assignee: LNK Corporation
• Inventors: Goodkovsky, Vladimir A.
• Primary Examiner(s): Davis, George B.
• Application Number:
  US09/801,059
• Publication Number:
  US 20020107681A1
• Time in Patent Office: 1,321 Days
• Field of Search: 706/3, 706/4, 706/5, 706/45
• US Class Current: 706/3
• CPC Class Codes: G09B 7/04 characterised by modifying the teaching programme in response to a wrong answer, e.g. repeating the question, supplying a further explanation

Abstract

A computer implemented method and apparatus for simulating an intelligent tutor for interactive adaptive training of learners in any domain includes a Domain Module, a Tutor Module and an Interface. Items to be learned, and their prerequisite and other dependency relationships are represented in a fuzzy graph, together with a fuzzy logic computational engine, which dynamically adapts the available sequence of training actions (such as presentations/explanations, simulations, exercises and tasks/questions) to a current assessment of the learner'"'"'s knowledge skill, the level of difficulty of the presented material, and preferences and learning style of the individual learner. Fuzzy logic is used as the basis of arc weightings, and the computations, but the general methodology is applicable to other approaches to weighting in computation.

Claims

1. An intelligent tutoring system for adaptive training of a learner in a domain comprising a field of knowledge, said tutoring system comprising:

a computer having an input for entry of information, a memory for storing information, a CPU for executing programs and an output for presentation of results of execution of a program; and

a computer readable memory encoded with a program for causing said computer to perform interactive training of said learner, said program including,

a Domain Module containing information concerning the field of knowledge, to be conveyed to the learner; and

a Tutor Module for simulating human-tutor training actions in testing, delivering information, practice and remediation for said learner using information selected and communicated from said Domain Module;

wherein said Tutor Module dynamically adapts a sequence of said training actions and selection of said information from said Domain Module, to a current assessment of the learner'"'"'s knowledge and skill level within the domain, utilizing principles of fuzzy logic.

2. An intelligent tutoring system according to claim 1, wherein adaptation of said sequence of training actions takes into account a level of difficulty of information to be presented and preferences and learning style of the learner.

3. An intelligent tutoring system according to claim 1, wherein said Domain Module comprises a reusable generic domain shell which is independent of said domain, for organizing domain data and interactive domain simulation.

4. An intelligent tutoring system according to claim 3, wherein said domain shell comprises:

a domain database defining a generic domain structure adapted to receive entry of domain specific data;

a domain player defining generic domain procedures; and

an interface for communicating information to and from the learner, an author and the Tutor Module.

5. An intelligent tutoring system according to claim 4, wherein said domain database stores domain data, comprising:

a plurality of domain objects adapted to interact with each other, with domain agents, with a learner and with the Tutor Module;

a plurality of domain agents for solving specific tasks by interacting with each other, with domain objects, with a learner and with the Tutor Module; and

an interface for communication of information among domain objects, domain agents, a learner and the Tutor Module.

6. An intelligent tutoring system according to claim 4, wherein the domain player comprises:

players for objects and domain agents;

a generic inter object interface to be filled in with specific interface from the domain database;

a generic domain learner interface, to be filled in with specific interface from the domain database;

a runtime manager; and

a timer.

7. An intelligent tutoring system according to claim 4, wherein said Domain Module further comprises a domain authoring tool for supporting input of a domain specific data into the domain database.

8. An intelligent tutoring system according to claim 7, wherein said domain authoring tool comprises a plurality of media editors for entry and editing of existing domain content in media form.

9. An intelligent tutoring system according to claim 1, wherein said Tutor Module comprises a reusable model-based tutor shell which includes predetermined training paradigm/domain/learner-generic pedagogical knowledge and skills, which is adaptable to a particular audience of learners and to use with a particular Domain Module, by the entry of data by an author during an authoring phase.

10. An intelligent tutoring system according to claim 9, wherein said tutor shell comprises checking/prognosis/diagnosis based dynamic planning capability, which makes decisions concerning delivery of training actions, based on said predetermined knowledge and skills, and said data entered by said author.

11. An intelligent tutoring system according to claim 10, wherein said tutor shell comprises:

a tutor database for defining and organizing a domain-independent generic tutor data structure and tutoring media data and metadata;

a tutor player for defining generic tutoring procedural modes and for implementing tutoring simulation; and

an interface for communication with an author, a learner and said Domain Module.

12. An intelligent tutoring system according to claim 11, wherein said tutor database comprises:

an audience database for providing a knowledge of a target audience;

a job/tasks database for supporting interaction with said author for training objective definition;

a media database for storing information entered by the author to provide the Tutor Module with paradigm/domain specific pedagogical knowledge; and

a learner database for storing information entered by the author for providing the Tutor Module with knowledge concerning specific learners.

13. An intelligent tutoring system according to claim 11, wherein said tutor player comprises:

a behavior recognition component which compares actions of a learner entered via said interface, with predefined patterns, selects a closest match, and generates performance pattern ID'"'"'s based on said closest match;

a cognitive interpretation function for recognizing performance pattern ID'"'"'s of a learner, using a cognitive interpretation model;

a decision making component for making decisions about sufficiency of training in different learning modes, and deciding which training mode to enter next;

an actions planning component for planning delivery of additional training actions; and

an action generation unit for generating a next training action via said interface.

14. A method for adaptive training of a learner in a domain comprising a field of knowledge, using a computer having an input for entry of information, a memory for storing information, a CPU for executing programs and processing information, and an output for presentation of results of execution of a program and processing information, said method comprising:

storing in said memory a Domain Module having a reusable generic domain shell which is independent of said domain for organizing domain data, and for interactive domain simulation;

storing in said memory a Tutor Module having a reusable model-based tutor shell which includes predetermined training paradigm/domain/learner-generic pedagogical knowledge and skills, and which is adaptable to a particular audience and a particular domain through entry of specific data;

entering domain specific, audience specific and learner specific data into said Domain Module and said Tutor Module, said audience comprising a group of prospective learners to be trained;

said computer performing interactive training of said learner based on information selected and communicated from said Domain Module and said Tutor Module, according to instructions from said Tutor Module.

15. A method according to claim 14, wherein said Tutor Module dynamically adapts a sequence of training actions and selection of information from said Domain Module to a current assessment of the learner'"'"'s knowledge and skill within the domain.

16. A method according to claim 15, wherein adaptation of said sequence of training actions takes into account a level of difficulty of information to be presented and preferences and learning style of the learner.

17. The method according to claim 15, wherein said adaptation is based on analysis of the learner'"'"'s knowledge within the domain and an author entered hierarchical structure which includes a plurality of rules concerning prerequisites and degrees of mastery, based on an application of principles of fuzzy logic.

18. A method for adaptive training of a learner in a domain comprising a field of knowledge, said method comprising:

providing a component including a hierarchically ordered set of data comprising units of domain information concerning said domain;

providing a paradigm for delivery of said domain information to said learner according to a hierarchy of said set of data, said paradigm including a plurality of rules concerning prerequisites for each of said units of domain information, including mastery levels for other units of domain information; and

dynamically adapting a sequence and manner of delivery to said learner of said units of domain information within said hierarchy, to a current assessment of the learners knowledge and skill level within the domain, based on an application of rules contained in said paradigm utilizing principles of fuzzy logic.

19. A method for operating a computer system for training a learner, the method comprising the steps of:

storing a tutoring module and a Domain Module in said computer;

associating the tutoring module and the Domain Module to obtain at least one learning paradigm having at least one training goal;

providing for interaction by the learner with the Domain Module, said interaction including said computer delivering training actions to said learner, and said learner entering responsive actions into said computer;

establishing a course for the learner comprising a sequence for delivery of training actions, based on the learning paradigm and the interaction by the learner; and

modifying the course based on the interaction by the learner to achieve at least one training goal of the learning paradigm.

20. A method according to claim 19, wherein said step of modifying the course comprises dynamically adapting said sequence of training actions to a current assessment of the learner'"'"'s knowledge and skill level within the domain, based on said responsive actions of the learner, using principles of fuzzy logic.

21. A method for instructing a learner using a computer, the method comprising the steps of:

providing a Domain Module associated with a Tutor Module thereby defining at least one training goal and at least one skill set related to each training goal via at least one teaching paradigm;

providing a learner interface to provide instruction and to allow the learner to interact with the Domain Module and the Tutor Module;

observing interaction between the learner and the Domain Module and the Tutor Module to obtain first interaction results;

comparing the first interaction results with the at least one skill set to obtain a skill level of the learner;

selecting a course of instruction based on the first interaction results and said at least one teaching paradigm;

providing active instruction to the learner based on the course of instruction;

monitoring the learner'"'"'s actions during the active instruction to obtain second interaction results;

modifying the course of instruction based on the second interaction results until at least one training goal is met;

providing passive instruction to the learner based on the course of interaction; and

providing the learner with control over tutoring modes.

22. A method according to claim 21, wherein said step of modifying the course comprises dynamically adapting said sequence of training actions to a current assessment of the learner'"'"'s knowledge and skill level within the domain, based on said responsive actions of the learner, using principles of fuzzy logic.

23. A method for operating a computer for training of a learner, the method comprising the steps of:

providing a Domain Module comprising media data;

providing a Tutor Module comprising at least one learning paradigm and at least one authoring tool, each learning paradigm associating media data of the Domain Module with at least one training objective and at least one skill set, each authoring tool defining for the Tutor Module a course module comprising audience data, job/task data, cognitive data, and training objectives, the audience data and cognitive data quantifying an educational goal, the course module being based on at least one relational association of course module elements to establish situated performance patterns;

providing the Domain Module data to the learner through an interface;

monitoring actions of the learner interacting with the Domain Module;

determining a level of knowledge of the learner for at least one skill set based on the actions of the learner interacting with the Domain Module;

selecting a training method based on the predicted level of knowledge of the learner;

generating at least one course module comprising a sequence of selected training actions, based on the level of knowledge of the learner;

providing instruction to the learner based on the course module;

monitoring progress by the learner through the course module;

identifying behavioral patterns of the learner during the progress of the learner through the course module;

updating the determined level of knowledge of the learner based on the identified behavioral patterns;

predicting a level of knowledge of the learner for at least one additional skill set based on at least one of the identified behavioral patterns, an associate of the skills sets, and the actions of the learner;

providing directed training of said course module based on the predicted level of knowledge of the learner for at least one additional skill set;

testing the level of knowledge of the learner to determine if at least one training goal is met; and

providing adjusted directed training through said course module until at least one training goal is set.

24. The method according to claim 23, wherein said step of providing adjusted directed training comprises dynamically adapting said sequence of training actions to the level of knowledge of the learner, using principles of fuzzy logic.

25. The method according to claim 23, further comprising the step of providing a pre-authoring phase.

26. The method according to claim 25, wherein the author analyzes the target audience to generate a database.

27. The method according to claim 26, further comprising the steps of:

generating an audience database and a job/task database;

obtaining correlation results between the job/task database and an audience profile;

generating job/task corrections based on correlation results;

defining training objectives; and

decomposing training objectives into sub-objectives.

28. The method according to claim 27, further comprising the steps of:

dividing sub-objectives into behavioral sub-objectives and cognitive sub-objectives;

obtaining correlation results between behavioral sub-objectives and job/task database;

obtaining correlation results between cognitive sub-objectives and the audience profile; and

generating sub-objective corrections based on correlation results.